Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    diamond problem

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    vide
    0.52
    avin
    0.50
     נוס
    0.48
    mor
    0.48
    pr
    0.48
    ica
    0.48
     ው
    0.48
    te
    0.47
     เพิ่ม
    0.47
    tra
    0.47
    POSITIVE LOGITS
     diálogo
    0.50
     estética
    0.49
    撤
    0.48
    冷
    0.47
     giữ
    0.46
     equipamentos
    0.46
    后端
    0.45
    布局
    0.45
     retiro
    0.45
    φέρ
    0.44
    Act Density 0.005%

    No Known Activations